Awesome

TIP Catalog for Apache Iceberg

This is TIP: A Rust-native implementation of the Apache Iceberg REST Catalog specification based on apache/iceberg-rust.

If you have questions, feature requests or just want a chat, we are hanging around in Discord!

Scope and Features

The Iceberg Protocol (TIP) based on REST has become the standard for catalogs in open Lakehouses. It natively enables multi-table commits, server-side deconflicting and much more. It is figuratively the (TIP) of the Iceberg.

We have started this implementation because we were missing customizability, support for on-premise deployments and other features that are important for us in existing Iceberg Catalogs. Please find following some of our focuses with this implementation:

• Customizable: Our implementation is meant to be extended. We expose the Database implementation, Secrets, Authorization, EventPublishing and ContractValidation as interfaces (Traits). This allows you to tap into any Access management system of your company or stream change events to any system you like - simply by implementing a handful methods. Please find more details in the Customization Guide.
• Change Events: Built-in support to emit change events (CloudEvents), which enables you to react to any change that happen to your tables.
• Change Approval: Changes can also be prohibited by external systems. This can be used to prohibit changes to tables that would invalidate Data Contracts, Quality SLOs etc. Simply integrate with your own change approval via our ContractVerification trait.
• Multi-Tenant capable: A single deployment of our catalog can serve multiple projects - all with a single entrypoint. All Iceberg and Warehouse configurations are completely separated between Warehouses.
• Written in Rust: Single 30Mb all-in-one binary - no JVM or Python env required.
• Storage Access Management: Built-in S3-Signing that enables support for self-hosted as well as AWS S3 WITHOUT sharing S3 credentials with clients. We are also working on vended-credentials!
• Well-Tested: Integration-tested with spark and pyiceberg (support for S3 with this catalog from pyiceberg 0.7.0)
• High Available & Horizontally Scalable: There is no local state - the catalog can be scaled horizontally and updated without downtimes.
• Openid provider integration: Use your own identity provider to secure access to the APIs, just set ICEBERG_REST__OPENID_PROVIDER_URI and you are good to go.
• Fine Grained Access (FGA) (Coming soon): Simple Role-Based access control is not enough for many rapidly evolving Data & Analytics initiatives. We are leveraging OpenFGA based on googles Zanzibar-Paper to implement authorization. If your company already has a different system in place, you can integrate with it by implementing a handful of methods in the AuthZHandler trait.

Please find following an overview of currently supported features. Please also check the Issues if you are missing something.

Quickstart

A Docker Container is available on quay.io. We have prepared a self-contained docker-compose file to demonstrate the usage of spark with our catalog:

git clone https://github.com/hansetag/iceberg-catalog.git
cd iceberg-catalog/examples
docker compose up

Then open your browser and head to localhost:8888.

For more information on deployment, please check the User Guide.

Status

Supported Operations - Iceberg-Rest

Operation	Status	Description
Namespace		All operations implemented
Table		All operations implemented - additional integration tests in development
Views		Remove unused files and log entries
Metrics		Endpoint is available but doesn't store the metrics

Storage Profile Support

Storage	Status	Comment
S3 - AWS		vended-credentials & remote-signing, assume role missing
S3 - Custom		vended-credentials & remote-signing, tested against minio
Azure ADLS Gen2
Azure Blob
Microsoft OneLake
Google Cloud Storage

Details on how to configure the storage profiles can be found in the Storage Guide.

Supported Catalog Backends

Backend	Status	Comment
Postgres
MongoDB

Supported Secret Stores

Backend	Status	Comment
Postgres
kv2 (hcp-vault)		userpass auth

Supported Event Stores

Backend	Status	Comment
Nats
Kafka

Supported Operations - Management API

Operation	Status	Description
Warehouse Management		Create / Update / Delete a Warehouse
AuthZ		Manage access to warehouses, namespaces and tables
More to come!

Auth(N/Z) Handlers

Operation	Status	Description
OIDC (AuthN)		Secure access to the catalog via OIDC
Custom (AuthZ)		If you are willing to implement a single rust Trait, the AuthZHandler can be implement to connect to your system
OpenFGA (AuthZ)		Internal Authorization management

Multiple Projects

The iceberg-rest server can host multiple independent warehouses that are again grouped by projects. The overall structure looks like this:

<project-1-uuid>/
├─ foo-warehouse
├─ bar-warehouse
<project-2-uuid>/
├─ foo-warehouse
├─ bas-warehouse

All warehouses use isolated namespaces and can be configured in client by specifying warehouse as '<project-uuid>/<warehouse-name>'. Warehouse Names inside Projects must be unique. We recommend using human readable names for warehouses.

If you do not need the hierarchy level of projects, set the ICEBERG_REST__DEFAULT_PROJECT_ID environment variable to the project you want to use. For single project deployments we recommend using the NULL UUID (" 00000000-0000-0000-0000-000000000000") as project-id. Users then just specify warehouse as <warehouse-name> when connecting.

Undrop Tables

When a table or view is dropped, it is not immediately deleted from the catalog. Instead, it is marked as dropped and a job for its cleanup is scheduled. The table, including its data if purgeRequested=True, is then deleted after the configured ICEBERG_TABULAR_EXPIRATION_DELAY_SECONDS (default: 7 days) have passed. This will allow for a recovery of tables that have been dropped by accident.

Configuration

The basic setup of the Catalog is configured via environment variables. As this catalog supports a multi-tenant setup, each catalog ("warehouse") also comes with its own configuration options including its Storage Configuration. The documentation of the Management-API for warehouses is hosted at the unprotected /swagger-ui endpoint.

Following options are global and apply to all warehouses:

General

Task queues

Currently, the catalog uses two task queues, one to ultimately delete soft-deleted tabulars and another to purge tabulars which have been deleted with the purgeRequested=True query parameter. The task queues are configured as follows:

The queues are currently implemented using the sqlx Postgres backend. If you want to use a different backend, you need to implement the TaskQueue trait.

Postgres

Configuration parameters if Postgres is used as a backend, you may either provide connection strings or use the PG_* environment variables, connection strings take precedence:

KV2 (HCP Vault)

Configuration parameters if a KV2 compatible storage is used as a backend. Currently, we only support the userpass authentication method. You may provide the envs as single values like ICEBERG_REST__KV2__URL=http://vault.local etc. or as a compound value like: ICEBERG_REST__KV2='{url="http://localhost:1234", user="test", password="test", secret_mount="secret"}'

Nats

If you want the server to publish events to a NATS server, set the following environment variables:

OpenID Connect

If you want to limit ac cess to the API, set ICEBERG_REST__OPENID_PROVIDER_URI to the URI of your OpenID Connect Provider. The catalog will then verify access tokens against this provider. The provider must have the .well-known/openid-configuration endpoint under ${ICEBERG_REST__OPENID_PROVIDER_URI}/.well-known/openid-configuration and the openid-configuration needs to have the jwks_uri and issuer defined.

If ICEBERG_REST__OPENID_PROVIDER_URI is set, every request needs have an authorization header, e.g.

curl {your-catalog-url}/catalog/v1/transactions/commit -X POST -H "authorization: Bearer {your-token-here}" -H "content-type: application/json" -d ...

License

Licensed under the Apache License, Version 2.0